Error detection in crystallographic models

Abstract

A variety of criteria were tested for identifying errors in protein crystal coordinates. Statistical analysis was based on comparisons of a highly refined crystal structure and several preliminary models derived from molecular replacement. A protocol employing temperature factors, real-space fit residuals, geometric strains, dihedral angles and shifts from the previous refinement cycle is developed. These results are generally applicable to the detection of errors in partially refined protein crystal structures.

Error Estimates of Protein Structure Coordinates and Deviations from Standard Geometry by Full-Matrix Refinement ofγB- andβB2-Crystallin

“…More recent attempts to estimate errors have included: the 'residue R factor' of Jones et al (1991); the tabulated R indices of Elango & Parthasarathy (1990); the refinement protocol of Carson et al (1994) which uses temperature factors, real-space fit residuals, geometric strains, dihedral angles and shifts from the previous refinement cycle; the 'discriminator' of Sevcik et al (1993) which assesses the likely errors on each atom in terms of its temperature factor divided by its electron density in the final 2IF,,[ -IF,.I map, or U/p where ,o is the electron density; an empirically derived six-parameter equation of Stroud & Fauman (1995), and the use of the diagonal elements of the inverse normal matrix in a final cycle of unrestrained least-squares refinement to give an estimate of the radial errors in atomic positions (Holland et al, 1990).…”

Prediction of Protein B-Factors Using Multi-Class Bounded SVM

“…Acc stand for the accuracy of each predictor, while CC stands for the Pearson correlation coefficient as defined in Eq. (5). The left graph shows the performance on training set, while the right graph shows the performance on test set.…”

“…For example, Bahar [4] proposed a Gaussian network model (GNM) to address this problem. It is well known that B-factor represents the extent of free movement of atom, and that higher atomic B-factor correlates with their exposed surface area [5]. As a result, Haliloglu [6] refined the GNM model and analyzed the vibrational motion of globular proteins.…”

Reliable quality-control methods for protein crystal structures

“…A noteworthy exception is a systematic study by Carson et al (1994) using ®ve validation criteria (temperature factor, real-space ®t, geometric strain, dihedral angle value and shift from previous re®nement cycle) and a data set of six structures. This work provided some useful information on the detection of gross structural errors but also revealed very different levels of usefulness in these criteria.…”